Image segmentation and three-dimensional reconstruction of the liver based on 64-slice spiral CT scanning data / 南方医科大学学报

ABSTRACT

<p><b>OBJECTIVE</b>To study the segmentation methods of the liver CT images and the value of 3-dimensional (3D) reconstruction of the liver in the planning of hepatic surgery.</p><p><b>METHODS</b>The 2D Digital Imaging and Communications in Medicine (DICOM) format data of the liver obtained from healthy volunteers were transformed into bmp format image, and the liver image segmentation was performed using Photoshop software. The 3D model was reconstructed using MIMICS software.</p><p><b>RESULTS</b>The DICOM format data of the liver obtained by 64 slice spiral CT included totally 658 slice images. The segmented liver image showed clear profiles and complete intrahepatic duct data were reserved. The segmented liver images were free of discontinuation during continuous observation. The liver surface and internal ductal system, including the hepatic arteries and veins, and the hepatic portal system and their branches, were represented clearly. The reconstructed liver allowed clear identification of the anatomic landmark and matched the actual liver volume. The reconstructed ductal structure were distinct and continuous with natural texture. The reconstructed liver and the hepatic internal duct system were simultaneously displayed by adjusting the transparency of the liver, and the blood vessels were also represented.</p><p><b>CONCLUSION</b>Segmentation of the liver images in different phases using Photoshop can be feasible for liver reconstruction. The reconstructed liver and the intrahepatic ductal structure allow vivid 3D observation of the spatial relationship among the major tracts and accurate estimation of the liver volume.</p>